Slashdot Mirror

You can probably check for Linux/BSD/Solaris drivers from Nvidia's website. The drivers you can find here: http://www.nvidia.com/content/drivers/drivers.asp

Currently they only have Windows XP/2000 drivers for it, but I would imagine that the Linux etc. drivers will be available soon.

Aren't the drivers for FreeBSD over a year old?

This and the lack of java is why I left FBSD behind.

It pisses me off.

Then I think, for most of your customers, the chipset is more important than the CPU. Both Intel and AMD make reliable and high performing CPUs. Until fairly recently, I think Intel chipsets (especially on Intel motherboards) have been more reliable than chipsets made for AMD CPUs. Of course, lame PC manufacturers can mess this all up with bad power supplies and motherboard designs.

I would think most of your customers that are asking for your advice would value reliability over performance, since current 64-bit Celerons and Semprons will likely have enough performance to last the life of their computers. For system reliability, I might recommend Intel from a good manufacturer and model line. The drawback of most low-cost retail Intel PCs is the lack of a PCI-Express x16 slot (although GMA 950 might be good enough), so buyers should shop carefully or get one built.

AMD offers significantly more performance per dollar, but buyers should be warned about common flakey chipsets. The newest chipsets from NVIDIA and ATI should be recommended.

A few months ago, I think a 64-bit Socket 775 Intel 945G chipset based platform would have been the best choice for many consumers. Today, an NVIDIA GeForce 6100 chipset based platform would probably be the better choice. This might change again when low-cost desktop Yonah PCs are released, again when Socket M2 Semprons arrive, again when Conroe-based Celerons arrive, etc.

I believe the point of that original statement was concerning the H.264 and DVD codec improvements compared to their main competitor, Nvidia. As you may or may not know, Nvidia charges money for their DVD codec:

http://www.nvidia.com/object/dvd_decoder.html

I will never buy another ATI product again because I've been burned so badly by drivers in the past - however, I'll give ATI credit for not following in the greedy footsteps of their competitor, Nvidia. Charging for a DVD codec that's optimized for their hardware is just stupid. It's not bad enough they charge hundreds of dollars for the hardware? I applaud ATI for this move.

• isn't free
• does much the same things as the new free ATI software described in TFA

> Ubuntu at least gets a clean install, though it fails to detect the ethernet connection...

Yippers. You're in luck, brother. I helped my fellow Ubuntian out the other month with this exact same problem. All you need to do is install the nForce chipset drivers, and you're good to go, g. Just search the forums for nForce and you'll start to froth at the mouth when you see the solution. Don't worry. It's a good kind of froth. In fact, your mouth will get so frothy you could even rub it on your cheeks and shave with it. That's just how good Ubuntu is my friend...

http://nvidia.com/object/LO_20020702_5500.html

Media and Communications Processor (MCP)

Part of the NVIDIA nForce and nForce2 platform processing architectures, the MCP replaces the "Southbridge" of traditional motherboard architectures. The MCP (including the MCP-D and MCP-T) delivers the most complete suite of integrated networking and communications devices including Ethernet, HomePNA 2.0, IEEE-1394a/FireWire(R) port, and up to six USB ports. In addition, the integrated audio processing unit (APU) provides support for Dolby(R) Digital 5.1 encoding.

First of all, that's bullshit. If binary driver modules can't be used with Linux, then what the fuck is this?! It must be a figment of my imagination, but I'm pretty damn sure it's an nVidia binary driver module for Linux!

Second of all, merely using binary drivers wouldn't cut it, because the NAZI* content companies insist on having a complete end-to-end DRM'd pathway. Effectively, this means closed-source software all the way. The whole system would have to be binary-only, from the kernel to the player application, and this is impossible with Linux. (And by "impossible," I mean that if you did do it, the result wouldn't be Linux anymore.)

*I don't mean that just as an epithet; they really are Fascist!

I'm essentially crypto ignorant. About all I've known to do was verify MD5 hashes on downloads. Now that this is by-and-large pointless, how to check the veracity of things like Linux ISO's, video drivers, etc, ad inifintum?

Press Release

Example of HDR/non-HDR lighting effect

Press Release

Example of HDR/non-HDR lighting effect

wow, for once there's a slashdot article i have insight on! (whether it's modded that way remains to be seen.... ;) )

i would actually be shocked if there weren't linux support. the ability to do what they want only need to be in the drivers. i've been doing a gpgpu feasability study as an internship and did an mpi video compressor (based on ffmpeg) in school. using a gpu for compression/transcoding is a project i was thinking of starting once i finally had some free time since it seems built for it. something like 24 instances running at once at a ridiculous amount of flops (puts a lot of cpus to shame, actually). if you have a simd project with 4D or under vectors, this is the way to go.

like i said, it really depends on the drivers. as long as they support some of the latest opengl extensions, you're good to go. languages like Cg and BrookGPU, as well as other shader languages, are cross-platform. they can also be used with directx, but fuck that. i prefer Cg, but ymmv. actually, the project might not be that hard, just needs enought people porting the algorithms to something like Cg.

that said, don't expect this to be great unless your video card is pci-express. the agp bus is heavily asymmetric towards data going out to the gpu. as more people start getting the fatter, more symmetric pipes of pci-e, look for more gpgpu projects to take off.

I have a 6600 GT (and Athlon 2400+) and it runs smoothly at 1024x768 with most of the settings dialed up.

Install the latest Doom 3 patch, the latest Nvidia Forceware patch, and the latest motherboard and sound card patches.

For me, turning off EAX, setting the card's Antialiasing and Anisotropic Filtering control panel to 'Application Preference', and setting image_cacheMinK higher in autoexec.cfg (see page 7&8 of the guide) all seemed to make the biggest difference.

For more info, try the tweak guide.

It's PCI-Express, or PCI-e. PCI-X is a high-speed extension to the parallel PCI specification used for high-end expansion cards (RAID controllers, raytracers, etc)

The lanes per board limit isn't imposed by PCI-e, it's imposed by the chipset. There are chipsets that have enough lanes to provide two x16 slots (and motherboards that use them to do this), however they are still somewhat on the high end.

Welcome to 2005. You do realize that there are alternatives to VIA now?

I have a decent solution that is quite cheap. half the cost of an ATI AiW. works great. no bs. I bought a PNY one but there are others available. It was from newegg but they don't have it anymore. ~$75.

Most video cards only do 32 bit "true color", that is 8 bits each for R,G,B and an 8 bit alpha. You won't see any difference between 8, 12 and 16 bit per channel images with most cards. You might have a fancier card with fancy drivers that are set well. Then you might be able to see the difference.

Nvidia 7800 has up to 128 bpp. So do other fancy cards.

My crummy nforce4 has no such options, even with the nvidia driver. This is no big deal to me now.

The differences between workstation and desktop graphics cards have virtually diminished. For the past several years, Quadros have used the same GPUs as the desktop GeForce line and and FireGLs have used the same RXXX chips the desktop Radeon line. They're the same architecture, and usually the driver set for one is indicative of the driver set for the other. Heck, for a long time Radeon users who wanted 3D acceleration in Linux actually would actually go to ATi's site and download the FireGL driver. Why ATi doesn't seem to want to just offer a Linux driver for Radeons tweaked for desktop application performance like nVidia's been doing for about four and a half years now is really kind of baffling.

You might try starting on nvidia's page.

Link.

If you scroll down there's some links to supported hardware.

Hope that's of some help.

I just got an OLED based VR visor from eMagin for $899, it has 800x600 resolution, headtracking, earbuds, and microphone built in. It was rumored for a while that it would be paired with the revolution for a complete VR experience, but that rumor has unfortunately since been dispelled. But I could definately see how the nintendo controllers could be paired with it to good effect.

At $900, the z800 is a big leap for VR in terms of quality. The closest LCD version sells for several hundred dollars more (many VR helmets costs many thousands of dollars) and is of lesser qaulity. It works very well with off the shelf first person shooters like HL2, Call of Duty, Doom and such and really adds a lot of depth to objects. It really helps, i mean both the wowness factor and being able to navigate and interact with close at hand objects.

At this point ATI doesn't support stereoscopic output, so you have to go with an nvidia card with their stereo 3d drivers.

Here is someone's (not mine) blog on what they have been doing with their eMagin 3d visor.

Maybe not ready for everyone, but certainly VR type equipment is starting to come back off the drawing board with many years of refinement and new technology being used to make the price and experience much better.

From TFA:

Newell was equally harsh, if not more so, on Sony for its design of the PS3 architecture and programming environment. "There are incredibly few programmers who can safely write code in the PlayStation 3 environment. And I totally see why Sony wants people to write code that runs on seven SPEs and a central processing unit, because that code is never going to run well anywhere else," he said.

What he seems to not understand/want to pretend isn't the case is the fact that the architecture of the Cell is a reflection of longstanding trends in computer architecture, not an exotic thing that Sony dreamed up to be troublesome.

In particular, there has been a longstanding disconnect between the growth in the amount of memory bandwidth available to chips versus the amount of computation that can be done on them. Computational capacity is growing much more quickly than memory access. Over enough years, this disconnect makes a big difference! Nowadways, processor architects will tell you that computation is basically free while communication is what is expensive.

Architectures ranging from GPUs to multicore CPUs to Cell take advantage of these trends in various ways, deliving much more computational capacity than standard CPUs. All of these architectures are deeply inherently parallel. There just isn't any other viable way to take advantage of all of this computation.

John Owens has a nice chapter in GPU Gems 2 on this topic.

If Newell (or whoever) doesn't want to program the SPEs on the Cell, he's free to just use the PPC CPU on it. And his game will be much slower than someone who uses it well. But there aren't going to be very many performance gains in the future to be had from single-threaded code running on CPUs. So while Cell is not trivial to program, none of the other choices are any easier. (Note that there are C/C++ compilers for the SPE instruction set, etc, so they're not *that* hard to program.)

(I'd like to hope that Newell actually knows all this and is just posturing in he middle of his Steam pimping and that this doesn't reflect reality in Valve's world!)

-matt

Rock on: NVIDIA_SoundStorm_Song.mp3

The problem with binary-only drivers is that older products can't be supported by the community, if/when the maker decides to stop supporting them. A real-world example is that Nvidia has stopped supporting older cards (do a search for "TNT"). I can understand not wanting to provide Linux drivers for the aged TNT series cards (despite owning one), but the original GeForce and the GeForce 2 aren't supported any longer either. I'd like to think there's some kind of technical limitation, but the realist in me thinks it's a marketing or cost-cutting measure.

Another problem is that binary drivers may be limited in terms of kernel support. Support for the 2.6 kernels would a must, but what about 2.4? How will the driver situation change when the 2.7/2.8 kernels are released?

Binary drivers are a "quick fix" for a problem, but not a long-term solution.

So what, exactly, is "easy to use" about Windows or other Microsoft products?

If your vendor does not supply a linux driver for your hardware, whose fault is that? It is not a problem with Linux. It is a problem with your hardware vendor. There are companies that supply linux drivers for their hardware. Hell, even Dell supplies linux drivers for some of their RAID controllers.

If your hardware manufacturer provided drivers for everything BUT Windows, would you complain about Microsoft or the hardware vendor?

Hooptie

Sorry, you don't need specialized hardware for motion blur. And I'm not quite sure what you mean by "universal processing units suck". Pixel shaders are pretty damn fast these days, and they're putting a lot more of them into GPUs.

There are plenty of good motion blur algorithms that work in 3D (with extrusion in the opposite direction of velocity), 2D post processing using shaders, and accumulation buffer techniques, all of which run well on modern hardware. Dedicated motion blur hardware would be redundant, and a total waste of die space.

Here's a few examples of motion blur algorithms for modern GPUs running on "universal processing units that suck":

http://download.developer.nvidia.com/developer/SDK /Individual_Samples/featured_effects.html (see the spinFX demo)

http://download.developer.nvidia.com/developer/SDK /Individual_Samples/samples.html (see "Motion Blur as a 2D Post-Processing Effect"

http://search.ati.com/nasearch.asp?Query=motion+bl ur&go.x=0&go.y=0&DefaultLanguage=16&Catalog=NASite &rdoCatalog=NASite&Start=&Total=&Stat=New (this is a search on "motion blur" on ati's developer page, there are several papers describing motion blur techniques)

Sorry, you don't need specialized hardware for motion blur. And I'm not quite sure what you mean by "universal processing units suck". Pixel shaders are pretty damn fast these days, and they're putting a lot more of them into GPUs.

There are plenty of good motion blur algorithms that work in 3D (with extrusion in the opposite direction of velocity), 2D post processing using shaders, and accumulation buffer techniques, all of which run well on modern hardware. Dedicated motion blur hardware would be redundant, and a total waste of die space.

Here's a few examples of motion blur algorithms for modern GPUs running on "universal processing units that suck":

http://download.developer.nvidia.com/developer/SDK /Individual_Samples/featured_effects.html (see the spinFX demo)

http://download.developer.nvidia.com/developer/SDK /Individual_Samples/samples.html (see "Motion Blur as a 2D Post-Processing Effect"

http://search.ati.com/nasearch.asp?Query=motion+bl ur&go.x=0&go.y=0&DefaultLanguage=16&Catalog=NASite &rdoCatalog=NASite&Start=&Total=&Stat=New (this is a search on "motion blur" on ati's developer page, there are several papers describing motion blur techniques)

Sorry, but not all of us will be able to afford to check out the fancy effects any time soon.

He was talking about developers. These days games seem to be all about graphics (sadly enough), but all the amazing graphics this technology has to deliver are of no use for game developers if they can't work with it. High profile engines like D3 and Unreal offer a full developement framework which makes development easier (and cheaper), covering pretty much everything a game developer could need. Including graphics.

    As for the sneak preview video itself... wow. It is mindblowing, and the motion blur effect alone makes the graphics much better. Still, i'd like to know what kind of hardware rendered that in realtime. And yes, for all we know, this is nothing more than a tech demo. An amazing tech demo, yes, but i've seen a lot of amazing tech demos elsewhere... i'll reserve judgment until the game is out.

Even 6200 series is going to get the feature (already got it?). why not more expensive/later model wouldn't? see it for youreself at nvidia's page

I have no idea if this help or not, but NVidia has a technology called ActiveArmor that may be of interest. In a nutshell, it's a Gigabit hardware firewall solution that is built into many inexpesive boards. Supposedly it can be used in both incoming and outgoing directions, allowing you to know immediately if a penetrator attempts to access improper network resources. Here's the schpiel:

ActiveArmor Firewall supports stateless and stateful inspection, Web-based management, pre-defined security profiles, port block filtering, remote administration, and provides an easy-to-use set-up wizard. In addition, ActiveArmor Firewall has anti-hacking features such as anti-IP-spoofing, anti-sniffing, anti-ARP-cache-poisoning, and anti-DHCP server-important security controls for corporate network environments. In a corporate setting, an end-point firewall (such as a desktop firewall) with anti-hacking capabilities can reduce the internally originated security breaches, and can inhibit desktops from generating unauthorized traffic. The result is improved overall security, with reduced requirements from the IT staff.

Again, I'm not sure if it's what you're looking for, but it's at least a very interesting product.

Then I'm at a loss. Here is what I could find:

I checked as of Linux kernel 2.0.35 this driver is fully supported. So unless you are using distributions from 10 years ago you should be fine. There is a standard diagnostic program for this network card which allows you see what's going wrong ne2k-pci.c so I'd consider that pretty well supported.

I've used the soundblaster 16 works fine. This also was addressed during the Linux 2.0 days. RedHat wrote some software to handle bugs google on "isapnptools" and check to make sure your setting are correct. Depending on your distribution there are getting passed to the sound module something like sb io=0x220 irq=7 dma=1 dma16=5 mpu_io=0x330 so you can search you /etc directory for a line that looks this.

Nvidia puts out a pretty good driver for their video cards. The GeForce 3 is right on their supported product list .

You are correct my last post was in error this is old hardware that should be fully supported and as far as I can tell every distribution supports this (though you may have install the Nvidia driver yourself since its nonfree). I'm not doubting you are having problems but it ain't driver problems or hardware support problems (at least if you have your hardware right). These are all fully supported and have been for a decade.

...that the greatest threat to Intel's market domination in the future is not going to come from AMD but from a company such as Nvidia.

Take a look at what they are doing with their GPUs right now and you can understand why someone would suggest this.

I don't really care how much slashdot fanboys rant about NVidia, the people who actually use high-end video cards in Linux know the truth - NVidia is and has always been oders of magnitude above the rest.

X != Linux

and not everyone uses X or Windows

http://www.nvidia.com/object/unix.html

Graphics Drivers
Linux IA32
Linux IA64
Linux AMD64/EMT64T
FreeBSD x86
Solaris x64/x86

nForce Drivers
Linux IA32 Drivers
Linux AMD64 Drivers

I am happy for you that *your* setup wins every time, mine's not listed.

That is why I only use NVIDIA video cards.

Actually NVIDIA's old cards that have been out for over a year support H.264.

http://www.nvidia.com/object/IO_16213.html

If you download The latest Windows Media Player and have a 6xxx series video card with "pure video" you can run hardware accelerated H.264 video.

Even the ultra-cheap 6200 line (which would work just fine in a mac-mini)can do this.

From NVIDIA's latest Linux driver changelog:

Removed support for legacy GPUs; please see "Chapter 1. Selecting and Downloading the NVIDIA Packages for Your System" in the README for details.

http://www.nvidia.com/object/linux_display_ia32_1. 0-7664.html

Here: http://www.nvidia.com/object/unix.html
FreeBSD and Solaris drivers for nVidia cards.
About performance: http://bulk.fefe.de/scalability/

Wasn't there still some problem like the DRI DRM not being ported or something?

DRI & DRM are not yet ported, but are being worked on. Without them you lose some hardware acceleration. (Though you don't need them for full hardware acceleration on nVIDIA cards if you use nVIDIA's recently released Solaris x86/x64 driver.)

This [link to Phil Brown's community agpgart driver] didn't work for me when I tried a few months back.

Solaris now includes its own agpgart driver, based in part on Phil's driver, as of the Solaris development build released today, and provides the source as well in OpenSolaris - see http://cvs.opensolaris.org/source/xref/usr/src/uts /i86pc/io/agpgart/

I note that the latest Nvidia drivers for Linux have added 'initial support for Xinerama + OpenGL' - in other words, I gather you can have a single OpenGL context spanned over multiple graphics cards.

See Appendix V in the drivers README - I haven't tried it, but it sounds like you'll be able to expand to three or more heads, so long as the resulting window is less than 4096 pixels across.

Any use? I've only got experience with OpenGL on a single, dual-head graphics card thanks to Twinview, but I have to admit that works brilliantly for me. Who knows what this new thing is like. :-)

porting Altivec heavy applications to MMX is going to be a pain

Honestly, I cant tell the difference that the additional 25 frames per second makes. If it were the difference between 5fps, and 30fps, *then* I would be able to understand.

Amen. PC graphics are ridiculously hyped.

"You're not a gamer unless you're playing with a 5900. You call yourself a gamer? Ha ha."
--http://www.nvidia.com/object/geforcefx_590 0_vide os.html (GeForce FX Experience)

However...

And what's the point of having 200+ FPS on a console when regular televisions (NTSC) can only process 29.99 frames per second?

Headroom. If you're getting 200+ fps on a tv, you could double the processing time per frame to make the game that much prettier and still get 100+ fps.

Also, TVs don't operate at 29.99 frames per second. The most correct number for modern color television is 60000/1001 (~59.94) fields per second. From a performance standpoint, it's a small technical detail since each field (odd or even lines) could be considered only half of a frame. However, compared to producing full frames, this display method halves the update latency of the display device.

That said, computer monitors typically display a full frame at a time and start at 60Hz refresh rates.

Why assume that all a gamer wants is graphics?

I'm firmly in the gameplay camp. I'd much rather have a fun game with good gameplay and plot (if applicable) than one with the very latest graphics. Still, there may be something to the continual push for prettier pictures.

I've noticed that I can still have a good time playing a game of (original Half-Life) Team Fortress with my friends. However, I once tried to get them to play one of my favorite games from the same era: Need For Speed 5. It was a total flop, and not because they didn't like racing games. Only a few people liked it, and those were the people who played it when it was new and the graphics were pretty good for their time.

By today's standards, older games (especially those of the 3D variety) look incredibly plain, which, IMHO, makes it harder to get into the experience for the first time. So while there may be no really good reason for the industry as a whole to keep pushing for glitzier graphics at the current rate, individual developers have one.

You call it "fragmentation", and I call it "diversity."

Playing DVD video has legal issues surrounding the implementation, and so it's still a huge problem. It's actually *illegal* in much of the world to distribute a turnkey system that will play commercial DVD's on Linux.

SP/DIF audio on my NForce board and my Delta cards is no problem at all -- and this is ALSA, something that is not dependent on your distribution at all.

I'd suggest that instead of waiting to get into linux, you'd just get a Knoppix or Ubuntu CD.

On the other hand, I'd also recommend Windows 2000 for a media PC.

You get into linux for A/V because you want to develop new stuff, or because you want to be involved in the development, or because you have requirements that are not met by other systems, not because you want the shortest path to an A/V system.

I'm into linux audio because I do my own synth software. I also do Windows.

I'm really happy that there are many, many different distributions. From business cards that can boot a machine and give you an ssh-enabled terminal, to user workstation installs, to systems optimized for audio, and lots more.

I don't think it is all that difficult to educate yourself on the differences, and you will quickly find that they are not so different.

I found Debian to be the most streamlined, easiest to maintain and by *far* the easiest to create and maintain the packages you develop or build yourself.

I'm more than sold on dpkg; my workplace uses dpkg to deploy everything, on all kinds of systems, not just Linux (mostly Solaris, in fact).

I just don't see how you could go wrong with a Debian-based system, to begin with. Although audio and video may be a pain regardless of what system you start with. Sometimes, a minor pain, and sometimes impossible, depending on your hardware.

With your Nforce2 chipset, you're in a pretty good situation, since it's one of the few audio devices that's expressly, actively supported by the manufacturer. http://www.nvidia.com/object/linux.html

Nvidia puts out

Nvidia puts out

Nvidia puts out

Don't forget NVidia and Nintendo as well. Got all the bases covered - sort of the gaming world's arms dealer...

Linux? Sure. The "PPE" portion of the Cell is a POWER64, which Linux already runs on. The "SPE" engines are effectively going to need their own kind of OS to manage them, but you could start with a mostly-user-space API and move it into the Linux kernel after people have figured out what that OS should really look like. This is all new stuff.

Looking at the CELL architecture overview, though, the Cell doesn't look to me like a desktop replacement. It looks like a video card replacement. Think about it: the biggest piece of closed-source, proprietary hardware in your PC right now is your video card, with its sekrit interfaces and binary-only drivers. We're already starting to see a movement towards more general-purpose use of that hardware with things like nVidia's Cg toolkit. The CELL is the logical next step in that direction. You'll have a video card that runs Linux (or, ideally, a video card that acts as just another (heterogeneous) processor in your system).

http://forums.nvidia.com/index.php?showtopic=3062 So I was right. It's still only a software issue. If you had spent just five minutes searching google instead of drooling on your keyboard while composing this incoherent rant, you hadn't made such a fool of yourself. Which most likely happens to you on a regular basis, and is probably the reason why you post anonymously.

I think this card is not (yet) for gamers. Or just very stupid gamers. :-) But this card is great for visualization and general purpose graphics development. You can't have enough RAM on a graphics card, if you are developing state of the art algorithms that run on the GPU. For example visualization of MRI data scales cubic (of course, because of its volumetric 3D nature). And with 512 MB RAM you can display much more detailed MRI images than before. Also more RAM is great for other GPU heavy algorithms. Just have a look at GPU Gems Part 1 and 2. Games, User Interfaces, Video, Applications, ... everything with images will get more and more GPU centric.